AMDA: Publications - Caring for the Ages

David Brechtelsbauer, MD, CMD, associate professor of family medicine at the University of South Dakota School of Medicine, Sioux Falls, and a member of Caring's editorial board, recently received a call from the infection control nurse at a Good Samaritan facility in Sioux Falls, S.D., where he serves as medical director. Would he evaluate one of the residents? Despite repeated courses of antibiotics, the resident's urine cultures still showed evidence of pyuria.

Following his examination of the resident, Dr. Brechtelsbauer surmised that the resident's elevated white blood cell counts were most probably not attributable to a urinary tract infection (UTI), but to a colonic-vesicular fistula secondary to her known colon cancer. Dr. Brechtelsbauer said he hoped the correct diagnosis of the resident's colonic fistula would "decrease the number of unnecessary and uninformative urine exams and the futile courses of antibiotic therapy."

A call and letter to the attending physicians, as well as direct communication with the patient and family (the daughter was present for the exam) led to a change in treatment plans.

This case illustrates several ongoing dilemmas associated with diagnosing and treating UTIs in the long-term care (LTC) setting. UTIs remain a crucial clinical issue for LTC practitioners, who must be vigilant about avoiding overuse of antibiotics, which puts residents at risk of developing antibiotic-resistant infections. At the same time, medical directors contend with economic pressures from facility administrators and pharmacists to select the most cost-effective treatments.

"Doctors are getting pressure not to treat too much, so that residents do not develop antibiotic-resistant infections," said Sefi Knoble, MD, CMD, medical director of Inglis House, Philadelphia, a skilled nursing facility specializing in the care of younger adults (average age 50) with disabilities. "But the nurses are getting pressure from families, residents, and sometimes [other] employees, who are saying, 'We have a problem; this person isn't right.'"

Certainly, because UTIs are the most frequent cause of sepsis in the elderly, active infections must not be overlooked. The difficulties come in determining which patients have active infections and which simply have bacteria in the urine. Most geriatric specialists now agree that asymptomatic bacteriuria need not be treated, but "with half of LTC residents suffering from dementia and therefore unable to cogently report symptoms," noted Dr. Brechtelsbauer, "how can one be confident that the resident is truly asymptomatic?"

Diagnostic complexities are often compounded for off-site medical directors, who rely on facility staff to conduct appropriate assessments and communicate changes in residents' status in a timely fashion. "There is a huge list of conditions that could make a person restless that day," explained Dr. Brechtelsbauer. "In some facilities, any time anyone does anything different, they get a course of antibiotics, which, of course, is not helpful in the larger scheme of things."

Medical directors can help improve the quality of UTI diagnosis and treatment by ensuring that facility staff members agree on objective criteria for suspecting a UTI and by fostering good communication between front-line caregivers, nursing staff, and attending physicians.

Which Signs Warrant Action?

Prompt diagnosis of a true UTI depends upon consideration of intrinsic as well as extrinsic factors. "There is no one simple way of evaluating someone," said Dr. Knoble. "Fever is not a reliable indicator in the elderly, and reporting of symptoms isn't reliable if they have dementia."

Physicians must be willing, she said, to consider an entire spectrum of factors: the health of the resident's immune system, their baseline bladder function, medications they take that can affect continence, and whether or not they have urinary retention. Although elderly patients often do not present with symptoms typical in a younger population, such as fever, dysuria, or flank pain, a change in mental status, decreased appetite, abdominal pain, new onset of incontinence or even respiratory distress can indicate a possible UTI.

Dr. Knoble often relies on results of urinalysis in her patients, the majority of whom are quadriplegic and who have Foley catheters due to neurogenic bladder. "The vast majority of our residents are colonized with bacteria," she said. "We have very few no-growth urine cultures at this institution compared to a traditional geriatric long-term care institution."

More telling of an incipient infection, she said, is the number of white cells in the urinalysis. Residents at Inglis House are less likely to report pain because they don't have sensation; or pain may present as a different symptom, such as agitation, restlessness, increased spasticity, or a change in their usual pattern of behavior. Fever in a resident with MS may be subtle--99.4--but if the person is also lethargic and not eating well, Dr. Knoble may start empiric treatment and order a urinalysis. Losing 24 hours waiting for results of a urinalysis can be more dangerous when the resident has MS, she said, because an infection is more likely to become a systemic infection in someone with a compromised immune system and neurogenic bladder.

Unlike Dr. Knoble's patients, who tend to be more vocal and in touch with their symptoms, elderly LTC residents with cognitive impairments are unable to report their own symptoms. Usually, frontline caregivers, such as nursing assistants, must discern changes in residents' behavior or status.

Barbara Resnick, PhD, CRNP, associate professor at the University of Maryland School of Nursing in Baltimore, finds that nurses with whom she works in LTC increasingly look for UTIs and have become more attuned to signs of agitation, confusion, and behavioral changes that can indicate infection in a resident who can't report their own symptoms.

An informal survey conducted by the National Association of Geriatric Nursing Assistants (NAGNA) confirms Dr. Resnick's observation. Lisa Cantrell, RNC, president and NAGNA co-founder, said that steering committee members participating in her survey reported that CNAs are often the first individuals in their facilities to observe and report signs of a UTI. The most common signs the CNAs reported were cloudy, dark urine; odorous urine; behavioral changes; uncommon confusion; urinating less frequently; stomachache; and loss of appetite.

Three researchers at the Weill Medical College of Cornell University in New York City tested an "illness warning instrument" they developed to assess behavioral and functional status changes as indicators of impending illness. In the study to validate the instrument, 23 nursing assistants used the 12-item checklist over a four-week period to notate changes in 74 nursing home residents.

Residents with an instrument-recorded change were more likely to develop an acute illness within seven days than those with no change. In addition, the study verified a 76% interobserver agreement between nursing assistants on the morning and afternoon shifts (Boockvar et al. Nursing assistants detect behavior changes in nursing home residents that precede acute illness: development and validation of an illness warning instrument. J Am Geriatr Soc. 2000; 48: 1086-1091).

Honing Reporting Skills

Dr. Brechtelsbauer agrees that the frontline caregiver, most often the CNA, is in the best position to observe patterns and changes. Medical directors should conduct in-services for CNAs regarding which symptoms are more suggestive of UTIs, such as:

A system for reporting such symptoms should also be in place, although the NAGNA survey, anecdotal reports, and a recent study by Longo et al (Barriers to timely care of acute infections in nursing homes: a preliminary qualitative study. JAMDA. 2002; 3: 360-365) indicate that reporting residents' acute symptoms and infections is often not guided by consistent procedures. In the Longo study, relaying information to physicians' offices occurred by phone and by fax, and physicians differed in their receptivity to either method.

Some physicians, including Dr. Brechtelsbauer, don't mind receiving a fax, while others do. "The custom in this community is that in most practices a fax is more welcomed than a phone call for non-emergent situations," he said. "The nursing home's problem is to figure out when a fax is appropriate and when a phone call would be better. This decision is influenced not only by the medical symptoms, but also depends upon the responsiveness of the [physician's] office."

"Whatever the form of communication, it's critically important for the nursing staff to make sure that a response is received and that action, where appropriate, is taken," said Dr. Knoble, who is troubled by the practice of notifying physicians via fax because sending a fax doesn't guarantee that the nursing staff will have an actual interchange with the physician.

"Sending a fax doesn't mean anything," she cautioned, "if the nursing staff does not get a response from the doctor. Someone with a change in status may not seem to have a non-emergent situation, but without prompt action it may become emergent really quickly."

Symptoms, such as those listed above, deserve immediate attention. A nursing home that uses faxes to communicate changing patient status should also have in place a fail-safe system for acceptable response times and a plan of action if a response is not received from the physician's office.

Dr. Brechtelsbauer advises nursing staff members to get specific when reporting signs and symptoms. Example: A statement, such as, "the urine looks dark today" or "the urine smells bad today" may or may not be helpful. Studies have shown, said Dr. Brechtelsbauer, that these conditions may be more predictive of under-hydration than of infection. But, he added, receiving any report, by fax or by telephone, is "better than not getting a report at all."

Jiska Cohen-Mansfield, PhD, ABPP, professor, Department of Health Care Sciences and of Prevention and Community Health at George Washington University, Washington, D.C., and director of the Research Institute at the Hebrew Home of Greater Washington, Rockville, Md., has explored the association between infection and agitation in nursing home residents with several colleagues (The impact of infection on agitation: Three case studies in the nursing home. Am J Alzheimer's Care and Related Dis Res. 1994; July/Aug: 30-34).

She agrees that if physicians receive reports of a physical or behavioral change in the resident they are "in a great place because then it's up to them to ask the next question. They [the physicians] already are getting a 'red light:' Something is happening. They know they have to start an investigation. It's not a problem that the nursing assistant didn't say more; it's great that they said something. The main part of our work is being a detective. We work together to continue this detective story."

When physicians want more specific information on a consistent basis, noted Dr. Cohen-Mansfield, they need to train frontline staff members concerning which signs to look for and which to report.

A nursing report of a change in urine should include other critical pieces of information, such as the patient's vital signs; whether they are already on antibiotics; whether a urine culture has been sent; and whether he or she has any allergies to antibiotics, said Dr. Knoble.

Timely & Cost-Effective Treatment

To treat acute, uncomplicated UTIs, the American Society of Consultant Pharmacists advises using amoxicillin or trimethoprim-sulfamethoxazole as first-line therapy (Judith L. Beizer. Urinary tract infections in elderly long-term care residents. www.ascp.com/public/pubs/cc/supp5/shtml).

Dr. Brechtelsbauer notes that if a patient is in obvious distress or discomfort, it may be reasonable to start the patient on an antibiotic while awaiting culture results. For non-catheterized patients, he often uses sulfamethoxazole-trimethoprim (Bactrim) or a high-dose amoxicillin. However, if he knows that the patient has had difficult-to-treat infections in the past, he might start with a quinolone. It is also appropriate to take into consideration non-microbiological issues, such as prescribing an antibiotic in liquid suspension for a resident who has swallowing problems. Sulfa drugs or quinolone remain Dr. Knoble's choices for first-line therapy as well.

If the patient has a catheter, Dr. Brechtelsbauer prefers to remove the older catheter, obtain a new urine specimen through a new catheter, culture that specimen, and treat based on culture/sensitivity results. This prevents treatment of residents merely colonized with bacteria. If the patient is too ill to wait until those results come back from the laboratory, Dr. Brechtelsbauer might use a quinolone until results are reported and then shift to another more specific antibiotic based on results.

Obtaining a urine sample can also be a challenge in elderly residents with dementia. Sampling urine from disposable diapers has been shown to be a reliable method in elderly incontinent women (J Am Geriatr Soc. 1993; 41:1182-1186), and work by Ouslander et al (An accurate method to obtain urine for culture in men with external catheters. Arch Intern Med. 1987; 147:286-288) in the late 1980s outlined an accurate method for obtaining urine from men for culture and sensitivity using a clean-catch technique combined with a fresh condom catheter.

Another important issue occurs "when the culture comes back in three days indicating that inexpensive antibiotics would be better than the super-antibiotic started empirically, " said Dr. Brechtelsbauer. "Will the physician have the courage to change that prescription? At that point, we could switch to the cheaper antibiotic, but that might be a difficult decision to make if the patient is doing all right on that drug."

Dr. Knoble recently treated a resident with recurrent Pseudomonas with a third-generation Cephalosporin for two weeks to try to avoid hospitalization. The resident also had functional decline, poor appetite, general malaise, and kidney stones. The infection was finally eradicated, but at quite a monetary cost. Her facility administrator pointed out that the bill for two weeks of intravenous therapy was almost $7,000. "These are tough decisions for nursing homes to make," she admitted.

ASCP advises that although oral quinolones can effectively treat complicated or recurrent infections (thus avoiding hospitalization), over-use of the agents can lead to resistant organisms. "Prevalence and sensitivity patterns of urine cultures in a specific institution can be useful when choosing empiric therapy for patients with their first UTI," noted Judith L. Beizer, PharmD, FASCP. "Pharmacists and physicians should request that the microbiology laboratory provide monthly reports, if possible. These reports are useful for infection control and can be used to develop algorithms for the treatment of UTIs and other infections in the institution."

Prevention & Monitoring

Dr. Resnick pointed out that simple UTI prevention measures can be instituted on an ongoing basis. Some measures address the risk factors inherent for these residents. For instance, to combat increased vaginal pH in postmenopausal women, intravaginal estradiol suppositories (Vagi-Fem) may be helpful. A simple tablet with a thin applicator, this product (already indicated for urinary incontinence) may also help reduce the incidence of UTIs.

Proper hydration, avoidance of constipation, and use of cranberry juice or tablets may also help. Drawing on one study of UTI prevention, Dr. Knoble reported that "supplementation with cranberry can be useful in preventing use of antibiotics in people with recurrent infections" (Are cranberry juice or cranberry products effective in the prevention or management of urinary tract infections? J Wound Ostomy Continence Nurs. 2002; 29:122-126).

One CNA in NAGNA's steering committee shared a strategy she used when two residents with UTIs resisted drinking sufficient water: She made use of the "100 Bottles of Beer on the Wall" song and enlisted the residents in the game of drinking down their glasses of cool water each hour.

At Inglis House, Dr. Knoble and others have formed an interdisciplinary taskforce to examine risk factors and prevalence of UTIs in catheterized patients. In addition, they will address the frequency of bag changes because this operation converts the closed system of a Foley catheter into an open one, introducing the risk of bacterial contamination. This issue promises to offer a challenge--not just to staff--but to residents, who often prefer more frequent bag changes.

The practitioners interviewed for this report agreed that it's not possible to discuss prevention techniques without addressing the harsh reality of staffing shortages. "It's a terrible thing to go home knowing that you should have given a resident more water or ambulated them more frequently," noted Ms. Cantrell. She and NAGNA are anticipating the approval of a new regulation that may alleviate some of the feeding workload for nursing assistants and allow them more direct care time with residents.

More Research Needed?

A dearth of new UTI clinical research regarding exists. We need additional research about benign urological conditions, said Monica Liebert, MD, director of the Office of Research for the American Urological Association. Dr. Liebert's office has worked closely with Congressman James Leach (R-IA) to introduce new legislation requiring the National Institutes of Health (NIH) to put an administrative structure in place to provide a urologic specialist at the NIDDK. This would ensure that research topics on non-cancerous urologic conditions, which nevertheless are costly to society and the health-care system, receive the national support and administrative guidance they deserve.

At a press conference held June 3 to introduce the Training and Research in Urology (TRU) Act, it was noted that spending for incontinence research costs only about 90 cents per patient, while costs of treating chronic incontinence total approximately $25 billion annually. Staggering, considering that incontinence is often the proximal cause of many nursing home admissions. The TRU Act, which contains authorization language and doesn't require additional NIH funding, will soon be introduced in the Senate, reported Dr. Liebert.

Lingering Thoughts

The bottom line about UTIs in LTC: Both physicians and nursing staff must recognize that they have a shared responsibility to deal with symptoms, said Dr. Resnick.

Have a list of symptoms that should be reported. "Physicians appreciate reporting information, because that's what they're depending on to make a decision," she said. "And those who have had a urinary tract infection know that they can be very uncomfortable. So the sooner you deal with it, the better."

Urinalysis & UTIs: What You Need to Know

Urinalysis (UA) remains among the most frequently ordered tests in the nursing home setting. Urinary tract infections (UTIs) are commonly treated or mistreated, depending on how well staff members understand and communicate UA results to the physician.

In general, urinalyses are obtained for:

Resident symptoms (irritation, burning, pain, change in frequency or character of urine);
Change in appearance of a resident's urine as noted by caregivers (foul odor, significant darkening, blood, cloudiness);
General resident decline, especially someone with poor historical abilities or previous history of bladder infections; and
Results gathered while monitoring the resident's ongoing chronic disease processes (i.e., diabetes and hypertension).

In the absence of other symptoms, simple dark or concentrated urine is not a good reason to order a urinalysis. Often, these symptoms are due to poor intake. Treat them by pushing fluids and observing the resident. Selectively obtain a urinalysis for general decline. Typically, the resident should experience an acute event, such as delirium, fever, and a sudden change in intake, in order to precipitate a urinalysis. When assessing a resident who may have a coincidental UTI, consider other causes of decline and investigate them.

Medicare requires specific diagnoses (ICD-9 codes) to order urinalyses. Normally, your lab will provide the codes if the requisition contains the symptoms or suspected diagnosis, such as urinary tract infection, painful urination, foul smelling or bloody urine, urinary incontinence, hypertension, or diabetes. Always include thorough documentation for tests in the resident's chart using short but complete descriptions.

Urinalysis Interpretation

Interpreting urinalyses is relatively straightforward. Basic information obtained is as below:

Specific gravity: This test reveals the concentration of urine most commonly elevated in dehydration; commonly low in fluid overloaded states.
PH: Elevation with proteus most commonly occurs with bacterial infection. PH is unreliable if urine is exposed too long before testing.
Nitrate: These breakdown products of bacterial infection are elevated in the presence of infection or in small amounts in asymptomatic bacteriuria.
Protein: Values are elevated in bladder infections, and in patients with diabetes, hypertension, or primary kidney disease.
Urobilinogen: Significant amounts of this form of bilirubin, which is excreted by the kidneys, indicate liver disease. However, a fair number of drugs interfere with urobilinogen and may give false positives.
Leukocyte esterase: Infection is almost always indicated with the elevation of this bacterial infection-related enzyme.
Casts: Nonspecific hyaline casts will be present; other casts may indicate glomerulonephritis/kidney inflammatory conditions.
Red blood cells (RBC): An elevation occurs in a number of problems, anywhere from kidney stones, cancers, infections, and polyps; presence indicates problems anywhere from the kidneys to the urethra. Small amounts (less than 10 RBC per high-powered field) are usually of limited concern in older patients.
White blood cell count (WBC): An elevation typically occurs during a UTI. Less than 10 WBC per high-powered field normally does not mean infection; greater than 10 WBC normally indicates infection.
Bacteria: A count of more than 1+ (or greater than 100,000 colonies on culture) indicates infection. Lesser amounts in the presence of WBCs or symptoms may warrant treatment.
Epithelial cells: The presence of these indicates contamination. They lie outside the urethra and their presence makes urinalysis suspicious for contamination and less reliable to judge infection.
Glucose: A spill occurs most commonly in diabetics with high sugar levels. Any significant amount is not normal.

The most important values in determining a possible UTI are nitrates, leukocyte esterase, white blood cells, and bacteria. Asymptomatic bacteriuria should not be treated and occurs when bacteria present without sufficient white blood cells, nitrates, or leukocytes to indicate a true infection. The physician should always know all four values in evaluating UTIs.

Your facility should have specific protocols for contacting the physician based on urinalysis results and the resident's condition. In general the physician should be contacted urgently when the patient is ill or if the urinalysis results are markedly abnormal. Fax results should include allergies and the resident's current antibiotics. Note lab results and physician response in the chart. Your documentation can be brief but should support the logic for the orders in the chart.

Source: Charles A. Crecelius, MD, PhD, FACP, CMD

The Ever-Present UTI

Ongoing challenges & solutions for optimal assessment & treatment in LTC